plot_ly() and ggplotly() functionsplot_geo()DataTableWe will work with the COVID data presented in lecture. Recall the dataset consists of COVID-19 cases and deaths in each US state during the course of the COVID epidemic. We will explore cases, deaths, and their population normalized values over time to identify trends.
## data extracted from New York Times state-level data from NYT Github repository
# https://github.com/nytimes/covid-19-data
## state-level population information from us_census_data available on GitHub repository:
# https://github.com/COVID19Tracking/associated-data/tree/master/us_census_data
# load COVID state-level data from NYT
### FINISH THE CODE HERE ###
cv_states <- as.data.frame(data.table::fread("https://raw.githubusercontent.com/nytimes/covid-19-data/master/us-states.csv"))
# load state population data
### FINISH THE CODE HERE ###
state_pops <- as.data.frame(data.table::fread("https://raw.githubusercontent.com/COVID19Tracking/associated-data/master/us_census_data/us_census_2018_population_estimates_states.csv"))
state_pops$abb <- state_pops$state
state_pops$state <- state_pops$state_name
state_pops$state_name <- NULL
### FINISH THE CODE HERE ###
cv_states <- merge(cv_states, state_pops, by="state")
head, and tail of the datadim(cv_states)
## [1] 12750 9
head(cv_states)
## state date fips cases deaths geo_id population pop_density abb
## 1 Alabama 2020-06-05 1 19387 676 1 4887871 96.50939 AL
## 2 Alabama 2020-07-29 1 83782 1538 1 4887871 96.50939 AL
## 3 Alabama 2020-08-12 1 104786 1882 1 4887871 96.50939 AL
## 4 Alabama 2020-05-12 1 10464 435 1 4887871 96.50939 AL
## 5 Alabama 2020-08-28 1 122185 2107 1 4887871 96.50939 AL
## 6 Alabama 2020-06-04 1 19072 653 1 4887871 96.50939 AL
tail(cv_states)
## state date fips cases deaths geo_id population pop_density abb
## 12745 Wyoming 2020-04-29 56 544 7 56 577737 5.950611 WY
## 12746 Wyoming 2020-07-30 56 2686 26 56 577737 5.950611 WY
## 12747 Wyoming 2020-10-05 56 6629 53 56 577737 5.950611 WY
## 12748 Wyoming 2020-06-02 56 912 17 56 577737 5.950611 WY
## 12749 Wyoming 2020-10-21 56 9848 61 56 577737 5.950611 WY
## 12750 Wyoming 2020-09-11 56 4264 42 56 577737 5.950611 WY
str(cv_states)
## 'data.frame': 12750 obs. of 9 variables:
## $ state : chr "Alabama" "Alabama" "Alabama" "Alabama" ...
## $ date : IDate, format: "2020-06-05" "2020-07-29" ...
## $ fips : int 1 1 1 1 1 1 1 1 1 1 ...
## $ cases : int 19387 83782 104786 10464 122185 19072 116710 144164 16032 11373 ...
## $ deaths : int 676 1538 1882 435 2107 653 2024 2437 583 483 ...
## $ geo_id : int 1 1 1 1 1 1 1 1 1 1 ...
## $ population : int 4887871 4887871 4887871 4887871 4887871 4887871 4887871 4887871 4887871 4887871 ...
## $ pop_density: num 96.5 96.5 96.5 96.5 96.5 ...
## $ abb : chr "AL" "AL" "AL" "AL" ...
After merging the covid-19 data and state population data, there are 12,630 raws and 9 coloumns in the merged dataset.
state and abb into a factor variable# format the date
cv_states$date <- as.Date(cv_states$date, format="%Y-%m-%d")
# format the state variable
state_list <- unique(cv_states$state)
cv_states$state <- factor(cv_states$state, levels = state_list)
# format the state abbreviation (abb) variable
### FINISH THE CODE HERE ###
abb_list = unique(cv_states$abb)
cv_states$abb = factor(cv_states$abb, levels=abb_list)
# order the data first by state, second by date
cv_states = cv_states[order(cv_states$state, cv_states$date),]
# Confirm the variables are now correctly formatted
str(cv_states)
## 'data.frame': 12750 obs. of 9 variables:
## $ state : Factor w/ 52 levels "Alabama","Alaska",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ date : Date, format: "2020-03-13" "2020-03-14" ...
## $ fips : int 1 1 1 1 1 1 1 1 1 1 ...
## $ cases : int 6 12 23 29 39 51 78 106 131 157 ...
## $ deaths : int 0 0 0 0 0 0 0 0 0 0 ...
## $ geo_id : int 1 1 1 1 1 1 1 1 1 1 ...
## $ population : int 4887871 4887871 4887871 4887871 4887871 4887871 4887871 4887871 4887871 4887871 ...
## $ pop_density: num 96.5 96.5 96.5 96.5 96.5 ...
## $ abb : Factor w/ 52 levels "AL","AK","AZ",..: 1 1 1 1 1 1 1 1 1 1 ...
head(cv_states)
## state date fips cases deaths geo_id population pop_density abb
## 105 Alabama 2020-03-13 1 6 0 1 4887871 96.50939 AL
## 222 Alabama 2020-03-14 1 12 0 1 4887871 96.50939 AL
## 86 Alabama 2020-03-15 1 23 0 1 4887871 96.50939 AL
## 134 Alabama 2020-03-16 1 29 0 1 4887871 96.50939 AL
## 31 Alabama 2020-03-17 1 39 0 1 4887871 96.50939 AL
## 17 Alabama 2020-03-18 1 51 0 1 4887871 96.50939 AL
tail(cv_states)
## state date fips cases deaths geo_id population pop_density abb
## 12559 Wyoming 2020-10-27 56 11806 77 56 577737 5.950611 WY
## 12586 Wyoming 2020-10-28 56 12146 77 56 577737 5.950611 WY
## 12560 Wyoming 2020-10-29 56 12507 87 56 577737 5.950611 WY
## 12547 Wyoming 2020-10-30 56 13028 87 56 577737 5.950611 WY
## 12630 Wyoming 2020-10-31 56 13298 87 56 577737 5.950611 WY
## 12684 Wyoming 2020-11-01 56 13723 87 56 577737 5.950611 WY
# Inspect the range values for each variable. What is the date range? The range of cases and deaths?
head(cv_states)
## state date fips cases deaths geo_id population pop_density abb
## 105 Alabama 2020-03-13 1 6 0 1 4887871 96.50939 AL
## 222 Alabama 2020-03-14 1 12 0 1 4887871 96.50939 AL
## 86 Alabama 2020-03-15 1 23 0 1 4887871 96.50939 AL
## 134 Alabama 2020-03-16 1 29 0 1 4887871 96.50939 AL
## 31 Alabama 2020-03-17 1 39 0 1 4887871 96.50939 AL
## 17 Alabama 2020-03-18 1 51 0 1 4887871 96.50939 AL
summary(cv_states)
## state date fips cases
## Washington : 286 Min. :2020-01-21 Min. : 1.00 Min. : 1
## Illinois : 283 1st Qu.:2020-05-02 1st Qu.:16.00 1st Qu.: 3452
## California : 282 Median :2020-07-02 Median :29.00 Median : 20942
## Arizona : 281 Mean :2020-07-01 Mean :29.77 Mean : 68322
## Massachusetts: 275 3rd Qu.:2020-09-01 3rd Qu.:44.00 3rd Qu.: 79560
## Wisconsin : 271 Max. :2020-11-01 Max. :72.00 Max. :957108
## (Other) :11072
## deaths geo_id population pop_density
## Min. : 0 Min. : 1.00 Min. : 577737 Min. : 1.292
## 1st Qu.: 79 1st Qu.:16.00 1st Qu.: 1805832 1st Qu.: 54.956
## Median : 548 Median :29.00 Median : 4468402 Median : 107.860
## Mean : 2326 Mean :29.77 Mean : 6557427 Mean : 420.731
## 3rd Qu.: 2308 3rd Qu.:44.00 3rd Qu.: 7535591 3rd Qu.: 229.511
## Max. :33174 Max. :72.00 Max. :39557045 Max. :11490.120
## NA's :234
## abb
## WA : 286
## IL : 283
## CA : 282
## AZ : 281
## MA : 275
## WI : 271
## (Other):11072
min(cv_states$date)
## [1] "2020-01-21"
max(cv_states$date)
## [1] "2020-11-01"
min(cv_states$deaths)
## [1] 0
max(cv_states$deaths)
## [1] 33174
min(cv_states$cases)
## [1] 1
max(cv_states$cases)
## [1] 957108
The variable ‘data’ ranges from 2020-01-021 to 2020-11-01; the number of deaths ranges from 0-33,174; the number of cases ranges from 1-957,108 through the time period. ### 4. Add new_cases and new_deaths and correct outliers
new_cases, and new deaths, new_deaths:
new_cases is equal to the difference between cases on date i and date i-1, starting on date i=2Use plotly for EDA: See if there are outliers or values that don’t make sense for new_cases and new_deaths. Which states and which dates have strange values?
Correct outliers: Set negative values for new_cases or new_deaths to 0
Recalculate cases and deaths as cumulative sum of updates new_cases and new_deaths
# Add variables for new_cases and new_deaths:
for (i in 1:length(state_list)) {
cv_subset = subset(cv_states, state == state_list[i])
cv_subset = cv_subset[order(cv_subset$date),]
# add starting level for new cases and deaths
cv_subset$new_cases = cv_subset$cases[1]
cv_subset$new_deaths = cv_subset$deaths[1]
#### FINISH THE CODE HERE ###
for (j in 2:nrow(cv_subset)) {
cv_subset$new_cases[j] = cv_subset$cases[j] - cv_subset$cases[j-1]
cv_subset$new_deaths[j] = cv_subset$deaths[j] - cv_subset$deaths[j-1]
}
# include in main dataset
cv_states$new_cases[cv_states$state==state_list[i]] = cv_subset$new_cases
cv_states$new_deaths[cv_states$state==state_list[i]] = cv_subset$new_deaths
}
# Inspect outliers in new_cases and new_deaths using plotly
### FINISH THE CODE HERE ###
p1<-ggplot(cv_states,
aes( x=date, y=new_cases, color=state )
) + geom_line() + geom_point(size = .5, alpha = 0.5)
ggplotly(p1)
p1<-NULL # to clear from workspace
### FINISH THE CODE HERE ###
p2<-ggplot(cv_states,
aes(x=date, y=new_deaths, color=state)
) + geom_line() + geom_point(size = .5, alpha = 0.5)
ggplotly(p2)
p2<-NULL # to clear from workspace
# set negative new case or death counts to 0
cv_states$new_cases[cv_states$new_cases<0] = 0
cv_states$new_deaths[cv_states$new_deaths<0] = 0
# Recalculate `cases` and `deaths` as cumulative sum of updates `new_cases` and `new_deaths`
for (i in 1:length(state_list)) {
cv_subset = subset(cv_states, state == state_list[i])
# add starting level for new cases and deaths
cv_subset$cases = cv_subset$cases[1]
cv_subset$deaths = cv_subset$deaths[1]
for (j in 2:nrow(cv_subset)) {
cv_subset$cases[j] = cv_subset$new_cases[j] + cv_subset$cases[j-1]
cv_subset$deaths[j] = cv_subset$new_deaths[j] + cv_subset$deaths[j-1]
}
# include in main dataset
cv_states$cases[cv_states$state==state_list[i]] = cv_subset$cases
cv_states$deaths[cv_states$state==state_list[i]] = cv_subset$deaths
}
Corrected outliers with number of new cases/deaths < 0. ### 5. Add additional variables
numeric). You can use the following variable names:
per100k = cases per 100,000 populationnewper100k= new cases per 100,000deathsper100k = deaths per 100,000newdeathsper100k = new deaths per 100,000Add a “naive CFR” variable representing deaths / cases on each date for each state
Create a dataframe representing values on the most recent date, cv_states_today, as done in lecture
# add population normalized (by 100,000) counts for each variable
cv_states$per100k = as.numeric(format(round(cv_states$cases/(cv_states$population/100000),1),nsmall=1))
cv_states$newper100k = as.numeric(format(round(cv_states$new_cases/(cv_states$population/100000),1),nsmall=1))
cv_states$deathsper100k = as.numeric(format(round(cv_states$deaths/(cv_states$population/100000),1),nsmall=1))
cv_states$newdeathsper100k = as.numeric(format(round(cv_states$new_deaths/(cv_states$population/100000),1),nsmall=1))
# add a naive_CFR variable = deaths / cases
cv_states = cv_states %>% mutate(naive_CFR = round((deaths*100/cases),2))
# create a `cv_states_today` variable
### FINISH THE CODE HERE ###
max_date <- max(cv_states$date)
cv_states_today = cv_states %>% filter(date==as.Date(max_date))
plot_ly()plot_ly() representing pop_density vs. various variables (e.g. cases, per100k, deaths, deathsper100k) for each state on most recent date (cv_states_today)
hovermode = "compare"# pop_density vs. cases
### FINISH THE CODE HERE ###
cv_states_today %>%
plot_ly(x = ~pop_density, y = ~cases,
type = 'scatter', mode = 'markers', color = ~state,
size = ~population, sizes = c(5, 70), marker = list(sizemode='diameter', opacity=0.5))
# filter out "District of Columbia"
cv_states_today_scatter <- cv_states_today %>% filter(state!="District of Columbia")
# pop_density vs. cases after filtering
cv_states_today_scatter %>%
plot_ly(x = ~pop_density, y = ~cases,
type = 'scatter', mode = 'markers', color = ~state,
size = ~population, sizes = c(5, 70), marker = list(sizemode='diameter', opacity=0.5))
# pop_density vs. deathsper100k
cv_states_today_scatter %>%
plot_ly(x =~pop_density, y = ~cases,
type = 'scatter', mode = 'markers', color = ~state,
size = ~population, sizes = c(5, 70), marker = list(sizemode='diameter', opacity=0.5))
# Adding hoverinfo
cv_states_today_scatter %>%
plot_ly(x = ~pop_density, y = ~deathsper100k,
type = 'scatter', mode = 'markers', color = ~state,
size = ~population, sizes = c(5, 70), marker = list(sizemode='diameter', opacity=0.5),
hoverinfo = 'text',
text = ~paste( paste(state, ":", sep=""), paste(" Cases per 100k: ", per100k, sep="") , paste(" Deaths per 100k: ",
deathsper100k, sep=""), sep = "<br>")) %>%
layout(title = "Population-normalized COVID-19 deaths (per 100k) vs. population density for US states",
yaxis = list(title = "Deaths per 100k"), xaxis = list(title = "Population Density"),
hovermode = "compare")
ggplotly() and geom_smooth()pop_density vs. newdeathsper100k create a chart with the same variables using gglot_ly()
geom_*() we need here?geom_smooth()
pop_density is a correlate of newdeathsper100k?### FINISH THE CODE HERE ###
p <- ggplot(cv_states_today_scatter, aes(x=pop_density, y=deathsper100k, size=population)) + geom_point() + geom_smooth()
ggplotly(p)
The population density of most states were concentrated below 400. Overall, a positive association between population density and number of new deaths per 100k was observed. The graph also showed a negative association when the population density was from 120 to 250.
naive_CFR for all states over time using plot_ly()
naive_CFR for the states that had a “first peak” in September. How have they changed over time?new_cases and new_deaths together in one plot. Hint: use add_lines()
# Line chart for naive_CFR for all states over time using `plot_ly()`
plot_ly(cv_states, x = ~date, y = ~naive_CFR, color = ~state, type = "scatter", mode = "lines")
# Line chart for Texas showing new_cases and new_deaths together
### FINISH THE CODE HERE ###
cv_states %>% filter(state=="Texas") %>% plot_ly(x = ~date, y = ~new_cases, type = "scatter", mode = "lines") %>% add_lines(x = ~date, y = ~new_deaths, type = "scatter", mode = "lines")
Zooming in on the interactive graph, peak value for number of new deaths is around July 27.
Create a heatmap to visualize new_cases for each state on each date greater than April 1st, 2020 - Start by mapping selected features in the dataframe into a matrix using the tidyr package function pivot_wider(), naming the rows and columns, as done in the lecture notes - Use plot_ly() to create a heatmap out of this matrix - Create a second heatmap in which the pattern of new_cases for each state over time becomes more clear by filtering to only look at dates every two weeks
# Map state, date, and new_cases to a matrix
library(tidyr)
cv_states_mat <- cv_states %>% select(state, date, new_cases) %>% filter(date>as.Date("2020-04-01"))
cv_states_mat2 <- as.data.frame(pivot_wider(cv_states_mat, names_from = state, values_from = new_cases))
rownames(cv_states_mat2) <- cv_states_mat2$date
cv_states_mat2$date <- NULL
cv_states_mat2 <- as.matrix(cv_states_mat2)
# Create a heatmap using plot_ly()
plot_ly(x=colnames(cv_states_mat2), y=rownames(cv_states_mat2),
z=~cv_states_mat2,
type="heatmap",
showscale=T)
# Create a second heatmap after filtering to only include dates every other week
filter_dates <- seq(as.Date("2020-04-01"), as.Date("2020-10-01"), by="2 weeks")
### FINISH THE CODE HERE ###
cv_states_mat <- cv_states %>% select(state, date, new_cases) %>% filter(date %in% filter_dates)
cv_states_mat2 <- as.data.frame(pivot_wider(cv_states_mat, names_from = state, values_from = new_cases))
rownames(cv_states_mat2) <- cv_states_mat2$date
cv_states_mat2$date <- NULL
cv_states_mat2 <- as.matrix(cv_states_mat2)
# Create a heatmap using plot_ly()
plot_ly(x=colnames(cv_states_mat2), y=rownames(cv_states_mat2),
z=~cv_states_mat2,
type="heatmap",
showscale=T)
naive_CFR by state on May 1st, 2020naive_CFR by state on most recent datesubplot(). Make sure the shading is for the same range of values (google is your friend for this)### For May 1 2020
# Extract the data for each state by its abbreviation
cv_CFR <- cv_states %>% filter(date=="2020-05-01") %>% select(state, abb, naive_CFR, cases, deaths) # select data
cv_CFR$state_name <- cv_CFR$state
cv_CFR$state <- cv_CFR$abb
cv_CFR$abb <- NULL
# Create hover text
cv_CFR$hover <- with(cv_CFR, paste(state_name, '<br>', "CFR: ", naive_CFR, '<br>', "Cases: ", cases, '<br>', "Deaths: ", deaths))
# Set up mapping details
set_map_details <- list(
scope = 'usa',
projection = list(type = 'albers usa'),
showlakes = TRUE,
lakecolor = toRGB('white')
)
# Make sure both maps are on the same color scale
shadeLimit <- 9
# Create the map
fig <- plot_geo(cv_CFR, locationmode = 'USA-states') %>%
add_trace(
z = ~naive_CFR, text = ~hover, locations = ~state,
color = ~naive_CFR, colors = 'Purples'
)
fig <- fig %>% colorbar(title = "CFR May 1 2020", limits = c(0,shadeLimit))
fig <- fig %>% layout(
title = paste('CFR by State as of', Sys.Date(), '<br>(Hover for value)'),
geo = set_map_details
)
fig_May1 <- fig
#############
### For Today
# Extract the data for each state by its abbreviation
cv_CFR <- cv_states_today %>% select(state, abb, naive_CFR, cases, deaths) # select data
cv_CFR$state_name <- cv_CFR$state
cv_CFR$state <- cv_CFR$abb
cv_CFR$abb <- NULL
# Create hover text
cv_CFR$hover <- with(cv_CFR, paste(state_name, '<br>', "CFR: ", naive_CFR, '<br>', "Cases: ", cases, '<br>', "Deaths: ", deaths))
# Set up mapping details
set_map_details <- list(
scope = 'usa',
projection = list(type = 'albers usa'),
showlakes = TRUE,
lakecolor = toRGB('white')
)
# Create the map
fig <- plot_geo(cv_CFR, locationmode = 'USA-states') %>%
add_trace(
z = ~naive_CFR, text = ~hover, locations = ~state,
color = ~naive_CFR, colors = 'Purples'
)
fig <- fig %>% colorbar(title = "CFR May 1 2020", limits = c(0,shadeLimit))
fig <- fig %>% layout(
title = paste('CFR by State as of', Sys.Date(), '<br>(Hover for value)'),
geo = set_map_details
)
fig_Today <- fig
### Plot side by side
### FINISH THE CODE HERE ###
subplot( fig_May1, fig_Today )
On May 1st, 2020, the highest CFR was observed on the east coast, with Connecticut (8.13) ranked the first and followed by New York (7.6). In California, the CFR is 4.1. Overall, the CFR across the nation decreased drastically from May to today, CFR equals to 6.38 in Connecticut and 6.47 in New York. It is 1.88 in California.